Labs · JanitorAI converter
JanitorAI bots and tavern cards,
translated both ways.
Paste a JanitorAI personality blob and walk out with a chara_card_v2/v3 JSON or PNG card — or drop a tavern card and get paste-ready text for every JanitorAI creation box, weighed against JanitorAI’s own 1,500-token guidance. What doesn’t fit gets listed, never silently dropped.
Conversion happens locally in your browser — nothing is uploaded or stored.
Fill at least Personality (or paste a blob) — the conversion appears here.
The field mapping, verified against JanitorAI’s own docs
The mapping below follows JanitorAI’s help-center description of the creation page (checked 2026-08; the page itself was last updated 2026-08-11) and the chara_card_v2/v3 spec as SillyTavern-family importers read it. It moves structure only — no field’s text is rewritten.
| JanitorAI box | Card field | Why |
|---|---|---|
| Character Name / Chat Name | name | The chat name (when present) becomes the card name, because {{char}} expands to it; the full title is kept in creator_notes. |
| Personality | description | Both are the main model-visible body. JanitorAI officially recommends putting traits, backstory and even example dialogue here. |
| Scenario | scenario | Same semantics on both sides — permanent context. |
| Initial Message | first_mes | The opening message, verbatim. |
| Example Dialogs | mes_example | Temporary-token examples; a <START> separator is added if missing, since tavern frontends split blocks on it. |
| Character Bio | creator_notes | JanitorAI's bio is display-only and never enters the prompt — exactly the creator_notes contract. |
| Tags | tags | Copied through; JanitorAI caps at 10, cards don't. |
What it can’t do (the honest part)
It can’t read hidden bots: JanitorAI keeps private definitions private, and this page only converts text you can already see. Going to JanitorAI, alternate greetings, embedded lorebooks, assets and in-card regex have no creation-page slot — each is reported as a loss with a suggested route, not dropped in silence. Token numbers are the same character-ratio estimate the rest of this site uses, good for budgeting against JanitorAI’s published thresholds, not for predicting their exact count. And system_prompt has no JanitorAI box at all — you decide whether it merges into Personality or your own Advanced Prompts.
FAQ
JanitorAI has no export button — what do I actually paste in?
Whatever you can see. For your own bots, open the creation page and copy each box (Personality, Scenario, Initial Message, Example Dialogs) — or paste the whole Personality blob into the big box here and the <character> / <scenario> / <START> containers get sorted automatically. For other people's private-definition bots, the definition is hidden by design and no converter can conjure it.
Why does JanitorAI's Personality become the card's description field instead of personality?
Because that's where the body text lives on each side. On JanitorAI, the Personality box officially carries "the majority of the bot's information" — traits, appearance, backstory, even example dialogue. On tavern cards that main-body role belongs to description; the tavern personality field is a short summary slot. Mapping body to body keeps the prompt weight where the author put it.
Where does the 1,500-token health check come from?
From JanitorAI's own help center: their token guide says well-made bots perform best under 1,500 permanent tokens and that past 2,000 you're "skating on thin ice"; the creation-page overview cites 2,500 as the ceiling. This tool weighs Personality + Scenario (the creator-controlled permanent block) with the same character-ratio estimate the rest of this site uses, and says clearly that JanitorAI's own tokenizer will differ.
What happens to the lorebook when I convert to JanitorAI format?
It's listed as a loss, not silently dropped. The JanitorAI creation page has no lorebook box — world info is a separate feature there, with real dialect traps (imported entries famously get stuck at 1% probability). Run the book through the Lorebook Migration Doctor before moving it.
Is anything uploaded?
No. Parsing, splitting, token estimates and PNG embedding all run in your browser. The page works offline once loaded.